Who should I follow to learn about MLOps from top experts?

If your goal is to **actually learn MLOps rather than just follow AI news**, I’d keep the list fairly small: - **Chip Huyen** - **#1 recommendation.** Excellent for ML systems design, deployment, serving, scaling, and the engineering tradeoffs behind production ML. Her work is especially good for understanding *why* production ML is difficult. [Real-time Machine Learning with Chip Huyen - Video | Agentic AI Foundation](https://home.mlops.community/public/videos/real-time-machine-learning-with-chip-huyen?utm_source=chatgpt.com) - **Eugene Yan** - Great for learning how ML systems work **at scale**, particularly recommender systems, experimentation, evaluation, and production architecture. [Recommendations | Hamel Husain](https://pages.hamel.dev/profile/recommendations?utm_source=chatgpt.com) - **Hamel Husain** - Very practical. Particularly valuable for evaluation, debugging, LLM systems, and the less glamorous parts of getting AI systems to work reliably. [Recommendations | Hamel Husain](https://pages.hamel.dev/profile/recommendations?utm_source=chatgpt.com) - **Emeli Dral** - Follow if you want to get strong at **ML observability, monitoring, drift, data quality, and production testing**. Evidently's free course is particularly useful. [Evidently AI - Open-source ML observability course](https://www.evidentlyai.com/ml-observability-course?utm_source=chatgpt.com) - **Goku Mohandas** - Strong choice if you prefer **hands-on, end-to-end projects** rather than primarily reading theory. - **Demetrios Brinkmann** - Follow for the broader MLOps practitioner ecosystem: interviews, talks, workshops, and discussions with people actually operating ML systems. [MLOps Community | Learn, Meet & Grow in Real-World MLOps](https://mlops.community/?utm_source=chatgpt.com) ### If I were starting today I'd use this combination: **Chip Huyen → Eugene Yan → Hamel Husain → MLOps Community** Then supplement them with: - **Full Stack Deep Learning** - excellent free material covering the lifecycle of AI products, including deployment and continual improvement. [The Full Stack](https://fullstackdeeplearning.com/?utm_source=chatgpt.com) - **DataTalks.Club / MLOps Zoomcamp** - particularly good for actually building pipelines, experiment tracking, deployment, and monitoring with tools such as MLflow, Prefect, Grafana, and Evidently. [MLOps Zoomcamp recap: how to monitor ML models in production?](https://www.evidentlyai.com/blog/mlops-zoomcamp-monitoring?utm_source=chatgpt.com) - **Evidently** - use it to learn the monitoring/observability side deeply. [Evidently AI - Open-source ML observability course](https://www.evidentlyai.com/ml-observability-course?utm_source=chatgpt.com) **One important distinction:** don't build your MLOps feed entirely around Kubernetes/cloud/tool announcements. The most valuable MLOps skill is learning to reason about the **entire ML lifecycle**: data → training → evaluation → versioning → deployment → monitoring → retraining. If you tell me whether you're **a beginner, ML engineer, data scientist, or software engineer transitioning into MLOps**, I can give you a much more targeted **10-person follow list + learning roadmap**.

**If you want to learn MLOps, follow practitioners who actively shape the field through open-source projects, education, and industry leadership. Some of the most influential voices include Matei Zaharia (Apache Spark, MLflow), Jeremy Lewi (Kubeflow), Abhishek Thakur (Kaggle Grandmaster, educator), Ville Tuulos (Metaflow), Chip Huyen (ML systems design), and Demetrios Brinkmann (community building).**[22 Notable MLOps Experts & Engineers (2026)](https://echoglobal.tech/technologies/mlops/)[Best MLOps Engineers You Can Trust on GitHub · GitHub](https://gist.github.com/lreverchuk/a71e9fc2311605929350125750b18087) ## 🌟 Key MLOps Experts to Follow Expert| Known For| Why Follow Them ---|---|--- **Matei Zaharia**| Creator of Apache Spark & MLflow| Deep insights into ML lifecycle management and scalable infrastructure. **Jeremy Lewi**| Co-founder of Kubeflow| Expertise in deploying ML workflows on Kubernetes. **Abhishek Thakur**| 4× Kaggle Grandmaster, author| Practical ML engineering tutorials and approachable education. **Ville Tuulos**| Co-creator of Metaflow| Focus on simplifying ML infrastructure for real-world teams. **Chip Huyen**| ML systems engineer, author| Sharp commentary on ML systems design and production challenges. **Hamel Husain**| Fastai, nbdev contributor| Automation and reproducibility in ML workflows. **Alejandro Saucedo**| Open-source MLOps tooling| Advocacy for ethical and transparent ML operations. **Josh Tobin**| ML observability tools| Guidance on monitoring and debugging ML in production. **Demetrios Brinkmann**| MLOps Community founder| Networking, podcasts, and grassroots knowledge sharing. **Noah Gift**| Educator, author| Practical MLOps education with a focus on cloud-native workflows. ## 📌 How to Engage With Them * **LinkedIn & Twitter/X**: Most share technical updates, blog posts, and event announcements. * **GitHub** : Follow repositories like MLflow, Kubeflow, Metaflow, Flyte, and W&B for hands-on learning. * **Podcasts & Blogs**: Demetrios Brinkmann’s _MLOps Community_ podcast and Chip Huyen’s essays are especially insightful. * **Books & Courses**: Abhishek Thakur’s _Approaching(Almost) Any Machine Learning Problem_ and Noah Gift’s cloud MLOps courses are practical starting points. ## ⚠️ Risks & Challenges * **Overhype vs. Reality**: Some voices emphasize cutting-edge tools that may not be production-ready. Balance learning with practical experimentation. * **Tool Fragmentation**: MLOps is evolving quickly; following multiple experts helps avoid tunnel vision on one stack. * **Community Bias**: Lists of “top experts” often come from vendors promoting themselves. Cross-check contributions on GitHub and conference talks for credibility.[Top MLOps Consulting Companies in 2026, Compared](https://winder.ai/top-mlops-consulting-companies-2026/) 👉 To tailor your learning, do you prefer **hands-on tutorials**, **thought leadership essays**, or **community discussions**? That will help me point you toward the most relevant experts and resources.

To master MLOps (Machine Learning Operations), following the right practitioners, researchers, and community leaders is one of the best ways to stay ahead of production patterns, tooling changes, and systems design principles. Here are the top thought leaders, creators, and engineers you should follow: ### 1\. Thought Leaders & Authors * **Chip Huyen:** Co-founder of Claypot AI and author of _Designing Machine Learning Systems_. She is one of the definitive voices on production ML, and her writing and lectures break down real-world system architecture, data drift, and infra challenges exceptionally well. * **Goku Mohandas:** Creator of **Made With ML** , an incredible end-to-end MLOps repository and community. His tutorials are widely considered the gold standard for learning how to build, test, deploy, and monitor production-grade ML applications from scratch. * **Andrew Ng:** Founder of DeepLearning.AI and ML Engineering for Production (MLOps) Specialization. While broad, his focus on **data-centric AI** and systematic operational patterns is foundational for anyone starting out. [MLOps courses to take in 2023 - Evidently AI](https://www.evidentlyai.com/blog/mlops-courses#:~:text=ML%20Engineering%20for%20Production%20\(MLOps\)%20specialization%20from,production%2C%20*%20Deploying%20ML%20Models%20in%20production.) ### 2\. Infrastructure & Tooling Experts * **Demetrios Brinkmann:** Founder of the **MLOps Community** (the largest community of MLOps practitioners). He hosts the _MLOps.community Podcast_ , which features raw, technical conversations with engineers building ML systems at companies like Netflix, Uber, and Spotify. * **Hamel Husain:** An influential ML/AI engineer who frequently shares deep, practical advice on fine-tuning, evaluating, and deploying LLMs and machine learning infrastructure. His blog and social channels are packed with production tips. * **Alexey Grigorev:** Founder of **DataTalks.Club** and the popular _MLOps Zoomcamp_. His open-source, cohort-based approach is phenomenal for hands-on learners who want to code along with real infrastructure. ### Where to engage with them: * **LinkedIn / X (Twitter):** Chip Huyen, Hamel Husain, and Demetrios Brinkmann frequently share architecture breakdowns and industry trends. * **The MLOps Community Slack:** Joining this community gives you direct access to thousands of engineers solving production ML problems daily. Are you looking to learn MLOps from a software engineering background, or are you transitioning from a data science role?

To master MLOps (Machine Learning Operations), you want to follow practitioners who actively build ML infrastructure, write definitive guides, and run communities where real-world production problems are discussed. Top experts, creators, and resources to follow include: * ** _Chip Huyen_** – A leading voice in ML systems design. She previously worked on ML infrastructure at Netflix and NVIDIA, taught ML Systems at Stanford, and authored __Designing Machine Learning Systems__. Follow her [Chip Huyen's Website](https://huyenchip.com/) and her blog/social channels for deep architectural insights. [Chip Huyen](https://huyenchip.com/)[Real-time Machine Learning with Chip Huyen - Video](https://home.mlops.community/public/videos/real-time-machine-learning-with-chip-huyen) * **The MLOps Community** (led by Demetrios Brinkmann) – The premier global hub for practitioners. They host daily/weekly slack discussions, live-streamed AMAs, podcasts, and meetups. Check out the MLOps Community Website to plug directly into practitioner-led learning. * ** _Alex Merkulov_ & The Marvelous MLOps Substack** – Excellent for practical, hands-on architectural blueprints and tooling breakdowns. You can read current breakdowns on the Marvelous MLOps Substack. * **DVC (Data Version Control) / Iterative.ai Blog** – Essential reading for understanding data pipelines, experiment tracking, and data-centric MLOps patterns. Their open-source guides are top-tier and hosted on the DVC Blog. [The Best Mlops Blogs and Websites - Feedly](https://feedly.com/i/top/mlops-blogs) * ** _Hamel Husain_** – Renowned for practical advice on LLMOps, developer workflows for AI, and bridging the gap between messy notebooks and clean production code. He frequently shares actionable engineering patterns on his personal site and LinkedIn. Would you prefer to start with **foundational books and courses** , or are you looking for **hands-on code repositories and tool tutorials** (like Docker, MLflow, or Kubeflow) to get started?